Choice Set Formation Behavior in Selecting Travel Routes: Application of an Interactive Online Survey Platform

One of the major challenges associated with the analysis of route choice modeling is the formulation of the choice set of alternatives that may allow a relatively accurate prediction of demand for travel routes. The subset of route alternatives in the choice set should be relevant and feasible and include the attributes considered most by travelers when they choose a route. This research investigated the role and significance of route choice set formations with a focus on the perspectives of the modeler and of travelers. Revealed preference data were collected from Sydney, Australia, residents about their choice of route for their most recent commuting trip. The survey tool was programmed to use the Google Maps application programming interfaces to collect the route choice information, including the selected route and the set of routes that were considered. Three discrete choice models were used to investigate the traveler’s inclination toward certain attributes of routes, considering both car and public transit routes with the master choice set. The effect of possible bias generated because of the formation of route choice from the perspective of the modeler was also analyzed and presented with the results. The results show the intuitive signs of various attributes, with travel time being the significant factor for route choice. The difference between the choice sets considered by the traveler and by the modeler also suggests that those considered by the modeler possess enough variation to offer the possibility of better capturing important factors affecting route choice behavior.